play-create-share gameplatform for the ipadthe ipad’s touchscreen and acceleration sensor. we...

Master of Science in Computer ScienceJune 2011Alf Inge Wang, IDI

Submission date:Supervisor:

Norwegian University of Science and TechnologyDepartment of Computer and Information Science

Play-Create-Share Gameplatform forthe iPadExploring Game Creation Technique on a Touch Screen Platform

Fredrik Ludvigsen

TDT4900 - Computer andInformation Science, Master Thesis

Spring 2011

Play-create-share gameplatform for the iPadby

Fredrik LudvigsenAdvisor: Alf Inge Wang

Abstract

This report presents the implementation and analysis of a simple create-and-play game for the iPad. The game is a platformer with a level editor, utilizingthe iPad’s touchscreen and acceleration sensor. We believe the results arealso applicable to games, that have level editors, for other types of post-PCtablet devices (Samsung Galaxy Tab, Motorola Xoom etc.).

The analysis comprise a usability test with a questionnaire. 33 universitystudents and 4 children attended the test. The attendants were given severalopportunities to provide feedback as free text. The free text answers havealso been categorized and summarized.

The level editor provided a low entropy building scheme which was popularamong the test population. Multiple mechanisms were provided for scrollinginside levels and controlling the game character. The population was dividedin which mechanisms they learned to use, and which they preferred. Webelieve that this indicates that they did all contribute to a good game andeditor experience.

We also provide some general advice about GUI appearance on touch baseddevices. Many of our test attendants were dissatisfied with how well theycould understand the meaning of the menu icons. Some attendants suggestedthat on-screen text should be used more extensively, but we do not concludewhether an improved set of icons, more text, a demonstration or informationmode or some other, yet unknown strategy is the best way to handle theissue.

i

Preface

This report is one of the main products of my master thesis in computerscience at NTNU during spring 2011. I chose this project because I wantedto get some real experience developing games, and the iPad seemed like anappropriate platform for a small game. The iPad also provided me with anopportunity to learn a new programming language and an API that is muchwanted in the software development business.

In the end I had not learned a lot about the iOS API’s, but a lot about casualgames, indie games and game editors in general. The usability tests and thefollowing analysis was an eye-opener to me. I learned much about the designand creation of self-explaining interfaces. The truth is that there is no magicbullet solving this problem, but there is hard work, valuable experience and anever ending improvement process. If software were buildings, games wouldbe the kindergartens; you never know what to expect and there is no recipefor fun.

I would like to thank Alf Inge Wang for being my advisor during this project.He helped with directing the project, searching for relevant technical mate-rial, providing frequent feedback underway and reviewing the report, amongother things. He even conducted a few of the user tests, voluntarily, dur-ing his own spare time. I would also like to thank all my test subjects andeveryone else who helped me during the project.

Fredrik Ludvigsen

ii

Contents

1 Introduction 1

1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Mobile software . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Research 3

2.1 Task description . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2.2 Level sharing platforms . . . . . . . . . . . . . . . . . . . . . . 3

2.3 Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.4 Scope of the project . . . . . . . . . . . . . . . . . . . . . . . . 4

2.5 Focus of the project . . . . . . . . . . . . . . . . . . . . . . . . 5

2.6 Research method . . . . . . . . . . . . . . . . . . . . . . . . . 6

3 Prestudy 8

3.1 Related Research . . . . . . . . . . . . . . . . . . . . . . . . . 8

3.2 The iPad platform . . . . . . . . . . . . . . . . . . . . . . . . 9

3.3 Real-time Strategy Games . . . . . . . . . . . . . . . . . . . . 11

3.4 Simulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3.5 Platformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

3.6 Puzzle Games . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.7 Social . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

iii

CONTENTS CONTENTS

3.8 Racing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

4 Own contribution 22

4.1 Initial System requirements . . . . . . . . . . . . . . . . . . . 22

4.2 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

4.3 Design & Arch. . . . . . . . . . . . . . . . . . . . . . . . . . . 24

4.4 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . 26

5 Results and evaluation 38

5.1 Experience with development tools . . . . . . . . . . . . . . . 38

5.2 User tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

5.3 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

6 Conclusion 58

6.1 Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

6.2 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

6.3 Further work . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

A Test document 62

B Sketches of levels created by the test attendants 71

C Raw answer data from the questionnaire 75

References 91

iv

Chapter 1

Introduction

1.1 Motivation

Since Android and iPhone smartphones started hitting the market in 2008and 2007 there has been a growing interest in games and other software fortouchscreen based devices. These devices cannot use traditional PC-basedsoftware for many reasons.

• They typically run on battery power.

• They have small screens.

• They typically have no keyboard or mouse.

Tablet computers has been available since long before Android and iPhonewere introduced, but recently a new category has been introduced. Devicesin the new category uses smartphone operative systems like Android andiOS. They are distinctively different from the previous tablets because theylack a pen-like input device. Older operative systems relied on the preciseinput of a mouse and were relatively hard to navigate only by touching thescreen. Throughout this report we will refer to the new tablet category aspost-PC tablets.

The goal of this project is to explore techniques that are suitable for imple-menting a game with an editor on a post-PC tablet.

1

1.2. MOBILE SOFTWARE CHAPTER 1. INTRODUCTION

1.2 Software for mobile operative systems

Dependency on a keyboard and mouse is typical for almost all software thathas been developed for PCs. Some software also uses the CPU extensively.Application candidates that could benefit from a transition to smartphonesand post-PC tablets should:

• use little CPU power

• function properly without large screen interfaces

• function properly without a keyboard

• function properly without right-click and mouse scrolling

Benefits from a transition to mobile devices includes:

• long battery life

• small size

• positioning technology

• cameras

• microphones

• multi-touch interfaces

• simple widespread market platforms

The post-PC tablet game market is dominated by casual games. Apple rec-ognized the state of their App Store and appropriately described it as a goldrush. But gold rushes end, and as they do, only the most adaptive golddiggers will remain—adapting appropriately to the platform will be key tothe post-PC tablet application market.

High quality applications like Garageband and iMovie demonstrates the con-venience of creating music and movies on iPad. This project will exploreoptions directed towards creating games on iPad, but with the limited re-sources of a single person for implementation and testing. Games like SecondLife, Little Big Planet and ModRacer for Playstation 3 has shown that har-vesting the creative power of users can result in games that renew themselves,prolonging their lives and giving them the time needed to become rich oncontent and loved by their users (Sanjeev et al., 2008).

2

Chapter 2

Research questions and method

2.1 Task description

Play-create-share is a new trend for video games, seen in gamessuch as Little Big Planet and ModRacer for Playstation 3. In thisproject, the goal is to develop a game with a game level editorfor the iPad. The project should utilize the iPad’s touch screento provide a new type of editor, encouraging the user to createlevels. The project aim for the development of a concrete gamethat has been designed by the supervisor.

2.2 Level sharing platforms

Jonathan Blow has shown examples of just how complex game developmentis (Blow, 2004). His article acted as a warning that high levels of complexitymight be one of the biggest challenges of any game creation project. By keep-ing the number of components low, we might have reduced the complexityexponentially. This is true since complexity emerges between the compo-nents. On average, for each component added, there is an increased chancethat new interactions between components are needed.

Since there was only a single person working on the project, no sharingplatform was developed for the game, but we made the game with such aplatform in mind, so that it would be easy to add in the future. We believethat sharing platforms are important for viral marketing simply because they

3

2.3. QUESTIONS CHAPTER 2. RESEARCH

create an incentive to spread the word.

2.3 Research questions

The research questions in this project are tied to the combination of play-create-share game editors and touchscreen interfaces. In the following listwe will abbreviate the concept of game level editors on post-PC tablets toml-editors, as in mobile level editors.

RQ1 How should developers address the lack of pointing devices like miceand pens in ml-editors?

RQ2 How should developers address the lack of traditional game controllersin ml-editors?

RQ3 How should developers address the need to edit levels that are biggerthan the screen in ml-editors?

RQ4 To what extent can ml-editors support creative processes?

For RQ1-3 a few different proposals have been implemented and tested. Thetest results could provide a few implications based on well the test attendantshandled simple tasks.

Through the test process, some examples of creative content (RQ4) have beencollected. These do not give any definite answers, but they might suggestthat the concept is worth looking into for other creative applications.

2.4 Scope of the project

This project was conducted as a part of video game research programme atNTNU, which started in 2008 (nor, 2007-2008). The only contributor to theimplementation code, except the libraries mentioned in Section 4.4 and a fewhundred lines of sample code, were the members of this project.

The scope included the development of a new type of game and an assessmentof its properties. It was important that the game would be fun to play, sowe focused on creating a small, but high quality game. By keeping the game

4

2.5. FOCUS OF THE PROJECT CHAPTER 2. RESEARCH

small, we left room for adjustments that could be needed in order to createan entertaining experience.

The following issues were relevant to the development:

• Game and editor mechanics in post-PC tablet games and editors

Relevant issues that did not receive substantial attention were:

• Aesthetics in post-PC tablet games

• Sound usage in post-PC tablet games

• Performance in post-PC tablet games

• Direct comparison of PC and post-PC tablet games

The game was tested by 37 people in order to assess the implications of thechoices that had been made during development. The test attendants weregiven questions about their gaming habits, their test performance and anyadvice they could be able to provide.

2.5 Focus of the project

RQ1 and RQ3 were core questions that could be relevant for any type of leveleditor on post-PC tablet platforms, while RQ2 was merely addressing gamesthat would have been played with joysticks and similar controllers on otherplatforms.

The creation of new game levels is a creative process. By assessing what kindof creative processes can take place between the level editor and the user,we would also assess why anyone would use it in the first place (RQ4). Theonly non-creative processes that could take place are the copying of otherlevels (completely directed) and entirely random productions (no direction).In both cases, the user will not have made anything that came from his orher own imagination.

The properties of mobile platforms, listed in Section 1.2, suggest that casualgames is a suitable genre for mobile platforms. This is also confirmed by themarket. Casual games like Angry Birds, Tiny Wings and Cut the Rope have

5

2.6. RESEARCH METHOD CHAPTER 2. RESEARCH

been among the most successful games at the Android Market and Apple AppStore.

(Taefay, 2010) lists four common elements that inform the design of casualgames:

• Rules and goal must be clear.

• Player need to be able to quickly reach proficiency.

• Casual game play adapts to a player’s life and schedule.

• Game concepts borrow familiar content and themes from life.

Reaching proficiency stand out because it is in direct conflict with the needto provide the player with powerful tools that enable creation of beautifuland challenging game levels.

Out of necessity, a lot of effort also went into learning the Objective-c pro-gramming language, the cocos2d game API and the Box2D physics API.

2.6 Research method

Basili has identified three common research approaches for software engineer-ing experiments (Basili, 1993):

The engineering method : By using the engineering experimental method,engineers build and test a system according to a hypothesis. Basedupon the result of the test, they improve the solution until it requiresno further improvement. The engineering method is typically used tofind better methods for structuring large systems, and software engi-neering is here viewed as a creative task not to be controlled by anythingelse than necessary restrictions on the resulting product.

The empirical method : A statistical method is proposed as a means tovalidate a given hypothesis. Unlike the analytical method, there maynot be a formal model or theory describing the hypothesis. Data iscollected to verify or falsify the hypothesis. The empirical method canbe applied on new technology to determine if this new technology isbetter or worse than the existing for producing software effectively.

6

2.6. RESEARCH METHOD CHAPTER 2. RESEARCH

The mathematical method : The mathematical method is based on math-ematical and formal methods for doing experiments. A formal theory isdeveloped and results derived from that theory can be compared withempirical observations. The mathematical method is usually used tofind better formal methods and languages, where software developmentis viewed as a mathematical transformation process.

The engineering method was used, but only for a single iteration. Here iswhat was done:

1. Literature study

2. Implementation of a platform game with a level editor for the iPad

3. A usability test with test subjects from the university. This included amodified version of the System Usability Scale (Lewis and Sauro, 2009).

4. Analysis of test results

Success criteria for this project relate to the creation of what has been de-scribed in Section 2.1 as "a new type of editor" for the iPad. Success in thisproject would call for an application that fits within any game definition andprovides an editor that could be used to create central parts of the gamecontent. Egenfeldt Nielsen, Heide Smith and Pajares Tosca have provided adiscussion of what is required for something to be classified as a game, but noabsolute answer is given, despite extensive efforts carried through by expertsin the field (Nielsen et al., 2008). Whether or not the game is successful inits mission, that is entertaining whoever uses it, is not a criteria, as anyonecould argue that a game is not entertaining.

7

Chapter 3

Prestudy - Games with editors

3.1 Related Research

In the specific area of games with editors on post-PC platforms we wereunable to find any research material. There is, however, lots of materialavailable about create-play-share games, fan based game modifications, gamecreation theory and other related concepts. In this chapter we will presentsome of them.

Second life is a popular MMO that provides what is called a metaverse,which are "fully immersive virtual spaces that significantly differ from onlinegames in several ways". In Second Life, sharing is a part of the performanceissue (Sanjeev et al., 2008). The amount of user created content provides achallenge for the implementation of the server. All of this content is collectedon the server and appears to players that are nearby. The content can alsoremain for a long time, which means that large amounts of content canaccumulate.

It has previously been shown how preference models can be used to createpersonalized game levels (Shaker et al.). These levels aim to optimize theexperience based on questionnaires administered to players after playing dif-ferent levels. While these kinds of levels are automatically created, the samestrategy might relate to user guidance in more classic editors.

It has previously been suggested that platform games are harder to produceby randomization strategies than RPG and strategy games (Compton andMateas, 2006). An algorithm for building platform patterns has been builtand tested. The algorithm takes platform jump distance and other kinds of

8

3.2. THE IPAD PLATFORM CHAPTER 3. PRESTUDY

difficulties into consideration, utilizing the rhythmic properties of platformergames. Rhythmic actions help the player reach states of higher concentration,called "flow" (Sweetser and Wyeth, 2005).

The legal issues of intellectual property rights to user created content havebeen looked into in (Humphreys et al., 2005). A case study of Auran andtheir 3D train and railroad simulator Trainz was carried through. The articleemphasized the importance of the question of which extent to which com-mercial entities can acknowledge fan based ownership as a port of the ’driftof value’ from producer to consumer. There is a balance between how muchfreedom is given to the users of an editor and how contractual restructuringis needed by the developer.

David Buckingham and Andrew Burn have discussed the need for a bettertheoretical game literacy framework in (Buckingham and Burn, 2007). Oneargument is the need for a better public understanding of games in order tocreate and use them for educational purposes. It is emphasized that in orderto teach through games, we should first teach about games, "If you wantto use television to teach somebody, you must first teach them how to usetelevision". This relates to games with editors because non-creative playingmight relate to reading the same way game level creation relate to writing.

Magy Seif El-Nasr and Brian K. Smith have presented experiences from at-tempts to teach students programming, mathematics, physics and aestheticprinciples by modifying existing games (El-Nasr and Smith, 2006). The au-thors observed that different game engines implicitly stress the use and devel-opment of certain skills, making the choosing of game engines an importantissue for the educational values. This represents a small step towards con-tent creation, and not only game "consumption". More importantly for thisproject, it demonstrates the great skill requirements of some powerful gamecontent editors.

3.2 The iPad platform

The iPad post-PC tablet was released on April 3, 2010. It is a 0.7 kg heavycomputer with a 242.8 mm by 189.7 mm touchscreen with multitouch sup-port. It is designed to be used with bared fingers, so non-conductive gloves orstylus pens does not work. It uses light sensors to automatically adjust thescreen brightness to its surroundings. Figure 3.1 shows Steve Jobs holdingan iPad.

9

3.2. THE IPAD PLATFORM CHAPTER 3. PRESTUDY

Figure 3.1: Steve Jobs holding an iPad

10

3.3. REAL-TIME STRATEGY GAMES CHAPTER 3. PRESTUDY

There is no intrinsic native orientation of the iPad, it responds to an accel-eration sensor by turning the interface so that it is in an upright position,unless overridden by the current application.

Apple claims that the iPad has 10 hours of battery when playing video. Thisis one of the properties that makes it more mobile than laptops.

Without modification it will only run software that either is approved byApple and distributed through the Apple App Store or runs in a web browser.

It supports 802.11a/b/g/n and Bluetooth 2.1 connectivity. The processor isa 1GHz Apple A4 custom-designed, high-performance, low-power system-on-a-chip with 256 MB DDR RAM (App, 2009).

3.3 Real-time Strategy Games

Many Real-time Strategy (RTS) games have level editors. Some examplesare Anno, Age of Empires, Warcraft, Starcraft and even Settlers.

Many RTS editors are powerful tools, including features like event controland the ability to create whole series of levels (El-Nasr and Smith, 2006).Starcraft and Age of Empires II have level editors that are independentprograms, derived from the tools that were used by the publishers to createsome of the original game content. Figure 3.2 shows the interface of the leveleditor which was included in the RTS game Starcraft.

The games work well on their own, but the editors extend their area ofapplication. For instance, a creative player can prepare a map to share withhis friends on a LAN party.

3.4 Simulators

The genre of simulator games includes a wide range of games. Some includelevel editors, but others do not. Some simulators even have level creationor other kinds of content creation as their main objectives. The first gamein the The Sims series could go on as long as the user wanted to. The userdecided when the creation had been completed or when it had been lost.

SimCity and The Sims allows the player to pause the game at any momentto edit the game world. Figure 3.3 shows The Sims when the simulation is

11

3.4. SIMULATORS CHAPTER 3. PRESTUDY

Figure 3.2: The level editor in Starcraft (Blizzard)

12

3.5. PLATFORMERS CHAPTER 3. PRESTUDY

Figure 3.3: The Sims

paused and building mode is active. Building material costs money, so theplayer always have to make sure there is a stable source of income available.This way the editing becomes more of an economic sustainability challenge,than a purely artistic challenge.

The Sims and SimCity are examples of games that focus on the creation ofnon-reusable content. The content they create does not have any use otherthan to continue building from a previous point in time or just displayingthe creation. Strict definitions of level editors might even exclude this kindof games because there is no difference between playing and level editing.

3.5 Platformers

Platformers that have level editors is a small, but diverse game category.The diversity might be linked to high level of entropy and a small set ofpossibly successful levels that the traditional platformers provide (Comptonand Mateas, 2006).

13

3.5. PLATFORMERS CHAPTER 3. PRESTUDY

Figure 3.4: Little Big Planet

Limbo is a game that is based on traditional 2D platformer concepts. Itsmind puzzling levels has brought the genre further, even without makingany big changes to it. This is an indication that platformer editors are toohard for amateurs to use. Another indication is presented in (Compton andMateas, 2006). Limbo came out in 2010, 25 years after Super Mario Bros.,but could still innovate, using the same concept.

Little Big Planet (LBP) andMinecraft demonstrate the diversity of the genre.LBP has a separate edit mode where the player is able to reverse time andundo creations freely. Minecraft has only one mode, where the player makesthe creations while playing the game. LBP is a 2D-physics based game whileMinecraft is based on discretized, approximate 3D-physics. Both games,however, rely on moving the character around while editing. Figure 3.4shows an image of the quasi-2d world of Little Big Planet.

Platformer editors are not as common as the real-time strategy level editorsand simulator editors, but a few games have become very popular. TheMinecraft project was started by a single developer, but became so popularthat he could expand the project even before the beta testing started. Thegame had grown large and profitable even with no money spent on advertising(Froholt, 2010). Figure 3.5 shows the blocky, discretized world of Minecraft.

14

3.6. PUZZLE GAMES CHAPTER 3. PRESTUDY

Figure 3.5: Minecraft

3.6 Puzzle Games

Puzzle games is a wide category for games that require careful thinking.

Crayon Physics is a puzzle game that allows the player to create shapes thatwill act physically within the game. This is the mechanism used both tosolve puzzles and to create them. The editor lets the user create puzzleswhile physics and switches are turned off.

To create a shape in Crayon Physics, the user must draw the shape with themouse cursor. This mechanism requires very little effort per created shape,and enable the player to build quickly. This is important because the playersometimes have to build while the game is running.

The Incredible Machine (TIM ) is similar to crayon physics, but is based ona discrete grid and a fixed set of objects that have complex interactions.Puzzles in TIM are solved while physics and other game object interactionsare halted, but with a fixed set of objects, set by the level creator. TheFigures 3.6-3.7 show TIM and Crayon Physics.

15

3.7. SOCIAL CHAPTER 3. PRESTUDY

Figure 3.6: The Incredible Machine

3.7 Social

The genre of social games is, not surprisingly, the dominating game genre inthe social network of Facebook. Social games often rely on viral marketing.Intrinsic high quality is a property that can help the viral marketing effectof any game (Sauro, 2011). For games with editors, this also applies to thecreations of the user. A player who is excited about his own creation mightbe one of the best ambassadors for the game.

For viral marketing to work well, the players should be able to share theircreations with other players (Taefay, 2010). Games like Farmville, CaféWorld and What To Wear use sharing intentionally to increase the amount ofplayers. The Figures 3.8-3.9 show images from What To Wear and Farmville.

Some of these games are strictly in the casual genre; they have more in com-mon with trophy shelves, chess clocks and paper based role-playing than withSimCity, Starcraft and The Incredible Machine, but they are very populargames based on the creativity of players.

16


Figure 3.7: Crayon Physics

17


Figure 3.8: What To Wear

18

3.8. RACING CHAPTER 3. PRESTUDY

Figure 3.9: Farmville

3.8 Racing

Few racing games include track editors, but there are some notable excep-tions. Trackmania might be the best known racing game with a track buildingsystem, but it was not the first. Stunts, which came out in 1990, includedloops, jumps and a track editor.

ModNation Racers represented a shift toward consoles. The editor hadchanged a lot, in order to be used with game controllers, but the groundprinciples were the same. The Figures 3.10-3.11 shows Stunts and Monda-tion Racers.

19


Figure 3.10: Stunts

20


Figure 3.11: ModNation Racers

21

Chapter 4

Own contribution

4.1 Initial System requirements

The initial system requirements of the system were derived from the taskdescription in Section 2.1 as following:

Pre-R1 It should be a game.

Pre-R2 It should be a platformer.

Pre-R3 It should include a level editor.

Pre-R4 It should produce content suitable for a sharing platform.

The concept of a game is hard to define. Some scientists say games should befun, some say they should be challenging and the list goes on with empiricaland subjective properties (Nielsen et al., 2008).

Instead of facing any definition of a game directly, we let the test subjectsdecide. They were told that they were going to test a game. Whether ornot the software was a game could, by our definition, be decided by theusers. Although this requirement is vague it is important, because it hasimplications on many of the properties of software.

We focused on making the game nice to look at, nice to listen to, fun andeasy to use (Pre-R1).

One proposed definition of platformer games is, "A 3D or side-scrolling 2Dgame in which a player must jump over/into/on objects/platforms in order

22

4.2. REQUIREMENTS CHAPTER 4. OWN CONTRIBUTION

to complete the game", (Answerbag, 2003). In other words, the game wasrequired to have platforms that the player could jump on and fall off of(Pre-R2). This implicated the existence of a character that the user control.

We chose to implement the level editor as a separate game mode (Pre-R3).It allowed the user to save the level at any time, and play it in play mode.

To enable easy sharing, we chose to save the levels in separate files (Pre-R4).There were no sharing platform created, but if one was built after the project,it would only have to handle one level per file.

Following is a summary of the refined system requirements:

R1 The system should satisfy the users’ definitions of a game.

R2-a The game should have platforms that the player can jump on and falloff.

R2-b The game should have a character that the user controls.

R3 The game should have at least two separate modes. At least one modeshould allow level modifications and at least one should not.

R4 Every game level should be stored in a separate file.

4.2 Mid Project System Requirements

We used the 2D-physics library Box2D for crash detection and gravity ren-dering. With this library, it was be easy to implement obstacles that behavedin a natural way. We decided to add some game objects that the charactercould interact with, as if crashing with them or pushing them.

We also wanted to use some classic platformer game mechanics, like points,a time constraint and start and end positions.

The following requirements were added during the project:

R5 The editor should allow the user to add objects that behave naturallyand which the user can interact with.

R6 The editor should allow the user to add objects that the character canpick up.

23

4.3. DESIGN & ARCH. CHAPTER 4. OWN CONTRIBUTION

R7 There should be a timing mechanism that is active while playing thelevels.

R8 There should be a start point and an end point that apply when playingthe levels.

4.3 Design & architecture

Implementation strategy

The combination of the cocos2d and Box2D libraries is common for iPhonegame development and it is easy to see why (Zyn, 2011), (Eri, 2011). cocos2dis a 2D-graphics library. It does not contain any specific game backbone,but a lot of tools common to 2D games. Menus, background music, spriterendering and display rotation were concepts that were easy to implementusing cocos2d. Box2D complemented this with a 2D physics simulation. Itdid not render anything on the display, but it provided interfaces that weregreat for implementing it.

Box2D supported 2D shape queries. This means it could be used to find outwhether there was any object present at any given point in the 2D space.Box2D would answer any space query with references to Box2D objects,which we would need to identify and associate with any texture that wewanted to use. As a convenience, Box2D objects contained an untyped ref-erence that could be used for any purpose. We used this reference to referobjects of our own class, that bound cocos2d objects, Box2D objects andapplication specific data together. Listing 4.1 shows the base class that wasinherited by all game objects.

Listing 4.1: Base class of game objects

@ in t e r f a c e GameObject : NSObject{

b2Body∗ body ; // Box2D s p e c i f i cCCSprite∗ s p r i t e ; // cocos2d s p e c i f i cCGSize bodySize ; // Appl i cat ion s p e c i f i c

}

24

4.3. DESIGN & ARCH. CHAPTER 4. OWN CONTRIBUTION

Grid based blocks

The grid based blocks were used in order to reduce the entropy of user createdcontent. This was one way of reducing, the set of choices that the user hadto make.

When two grid based blocks lined up, they were supposed to behave nodifferent from a single block covering the same area. By replacing all lined upblocks with an equivalent set of larger blocks, both the rendering algorithmsand the physics simulations would have fewer bodies to work with.

The problem of merging many blocks into as few bigger blocks as possible isan NP-hard problem (Lingas, 1982). Solving this problem was not necessary,but doing so could have increased the performance of the application afterthe merging was complete.

Pausing the application while building could degrade the experience of theuser, for instance by not responding appropriately to finger swipes. Swipingwith one or more fingers should produce series of blocks.

In order to increase the minimum performance of the editor we applied twoblock transformation restrictions.

• Blocks should only split when parts are being removed explicitly by theuser.

• Blocks should only merge after new blocks have been created explicitlyby the user.

With this restriction, the computational cost off adding a block would neverexceed O(n), where n is the number of existing blocks. Also the computa-tional cost of filling the interior of some empty area, would not exceed O(n),where n is the number of added blocks.

GUI design ideas

Scrolling

In the first implementation of the game, the levels had unlimited space inthree directions. The fourth was a limited by an infinitely wide floor object.It became apparent that the user would have to view a little bit at a time,

25

4.4. IMPLEMENTATION CHAPTER 4. OWN CONTRIBUTION

and we wanted this to feel natural. We chose to test three different mecha-nisms for sliding the viewing perspective horizontally, vertically or diagonally.Figure 4.13 shows what the mechanisms looked like.

Character control

We chose to use classic control mechanisms, known from Super Mario Bros.,to control the game character, but the iPad did not have built-in gamecontrollers or general purpose hardware buttons. We needed to control thehorizontal movement and jumping for the character. We decided to use anemulated joystick for horizontal movements and the rest of the screen surface,except for the menu area, for jumping. We also decided to allow the user tocontrol the horizontal movement by tilting the iPad. This was convenient,since an acceleration sensor is built into every iPad. Figure 4.12 shows thegame with the device tilting control enabled.

Menu appearance

The menu was built to be independent of language. It was based on glyphsthat the users were supposed to recognize the meaning of. There was, how-ever, a single line with information in English at the top of the screen thatcould help the user understand how to use every mode. For example the in-formation bar would display "Delete/Load/Save/Exit?" when the file menuwas active; Figure 4.7 illustrates this.

4.4 Implementation

Development process

The implementation of the game was a 4 month continuous developmentprocess with the goal of producing a game as described in Section 2.1 witha few modifications. The development method was incremental and basedon a minimal sample implementation of an iPhone game. The sample was ademonstration of how to create cocos2d + Box2D based project for XCode,the official iOS development environment, provided by Apple.

The system was tested and demonstrated every week and new goals weredefined based on the demonstration.

26


Platformer Playground

OpenGL ESBox2d

Cocos2DUIKit

Figure 4.1: Library stack

Objective-C++ implementation

Figure 4.1 shows a simplified overview of the libraries used in the project.Platformer Playground was the name of our game. Box2D is an open sourcephysics simulation library for 2D physics. It is written in C++, but therealso exists ports for other languages. The difference in the implementationof Box2D and cocos2d meant that the project would have to include bothC++ code and Objective-C code.

UIKit is the official UI library for iOS that has been provided by Apple.

cocos2d and UIKit contained some features that were similar. The UIKit pro-vided a lot of widgets that were familiar to iOS users, like the UIScrollViewand UIKeyboard, but it was not comprehensive enough for performance-intensive games. cocos2d and UIKit used different coordinate system han-dling routines, which forced us to provide our own implementation, just forsynchronizing them. For instance, the UIScrollView had to be handled care-fully by our code when the cocos2d library was performing interface rotationin response to rotation of the iPad device.

Figure 4.2 shows an overview our scene classes. The CCScene class, providedby cocos2d, was used to implement the top level, mutually exclusive gamemodes.

MenuScene was a small class that handled the menu layout on its own.It contained a text field and inherited the UITextFieldDelegate in order tocommunicate with the UIKeyboard. The PlayScene and BuildScene classesleft the details to other, more specialized classes, like GameWorld and Hud.

The Hud and PlayModeHud classes implemented the custom GUI widgets inthe editor and in the play mode. These are shown in Figure 4.3.

Figure 4.4 presents the GameWorld class. The GameWorld singleton wasthe owner of the game objects, which are presented in Figure 4.5. The

27


Page 1 of 1

Untitled 6/7/11 7:12 PM

CCScenePropertiesOperations

PlaySceneProperties

gw:GameWorld * hud:PlayModeHud * rotationLockDelegat…

Operations

BuildSceneProperties

gw:GameWorld * hud:Hud * rotationLockDelegat…

Operations

<UITextFieldDelegate>Operations

MenuScenePropertiesOperations

Figure 4.2: Scenes

Page 1 of 1


PlayModeHudPropertiesOperations

CCLayerPropertiesOperations

<ViewOffsetListener>Operations

HudPropertiesOperations

Figure 4.3: Widget overlay classes

GameWorld object handled adding and removing objects using functionsand objects from cocos2d, Box2D and a few custom classes, which have notbeen described in detail in this report.

Platformer Playground - The game

Figure 4.6 shows the main menu of Platformer Playground. Selecting thehammer symbol would cause a list of stored levels and a text field withthe placeholder ’<New playground>’ to appear. Entering any level namefollowed by ’return’ would add a new level to the list. Tapping one of thelevel names in the list would open the editor and start editing the selectedlevel.

Page 1 of 1


CCLayerPropertiesOperations

<GameEventDelegate>Operations

<HudDelegate>Operations

<JoystickDelegate>Operations

GameWorldPropertiesOperations

<UIScrollViewDeleg…Operations

Figure 4.4: Game object controlling class

28


Page 1 of 1


<Bouncable>Operations

bounce:(CGPoint)dire…

BouncerObjectPropertiesOperations

GameObjectProperties

body:b2Body * bodySize:CGSize sprite:CCSprite *

Operations

StartObjectPropertiesOperations

PlayerObjectPropertiesOperations

GoalObjectPropertiesOperations

GamePointlessPropertiesOperations

GamePointPropertiesOperations

NSObjectPropertiesOperations

GameBlockPropertiesOperations

BoundingObjectPropertiesOperations

Figure 4.5: Game Object classes

Figure 4.6: The main menu and the build level menu.

29


Figure 4.7: The editor when idle and when the file menu is activated.

The editor of Platformer Playground had a combined menu and toolbox thatwas based on states. Figure 4.6 illustrates following states:

• ’edit//’

• ’edit//file’

Figure 4.8 shows a complete tree of all states in the game. While using theeditor, there was only one menu available, residing in the upper right corner.This menu changed itself depending on what state the game was in. Forinstance it would only show options relevant to level modification while the’/edit//build’ mode was active.

In order to add or remove content, the user had to activate the buildingmode. Each placeable item had its own icon in the menu when build modewas active. The glowing icon at the top of the menu showed which editormode that was active; tapping it would deactivate that mode and activate’edit//’. Figure 4.9 shows the build mode.

The grass blocks were static building blocks. They behaved like infinitelyhard, floating objects with infinite mass.

30


build

gra

ss b

lock

point

pointless

bounce

rsp

aw

n p

oin

tgoal

era

se

file

save

load

dele

teexit

vie

wte

st

pla

y

Level 1

...

Level N

hig

hsc

ore

edit

Level 1

...

Level N

Level 1

...

Level N

Figu

re4.8:

The

states

ofPlatform

erPlaygrou

nd

31


Figure 4.9: Placing grass blocks, pointless and point

The point and pointless marbles were affected by physics simulation at alltimes. Placing a marble in mid-air would cause it to fall directly down onceit appeared, even if the player was neither in play mode or test mode.

Point marbles could be picked up by the player, either in test mode or in playmode. Pointless marbles looked a little different, and could not be picked upby the player, but were in every other way similar to point marbles.

Figure 4.10 shows the springs. The springs were static, grid based objects.Springs affected marbles and the player by giving it an impulse pointingstraight up. The character got a greater impulse. The idea was that theobject would bounce as much as possible, but not making it hard too for theplayer to play with. Because the rabbit character should be controlled by theplayer, giving it a powerful impulse would not confuse the player too much.Different impulse strengths were not tested after the initial implementation.

Figure 4.11 shows a level with a spawning point and a goal flag. These werenot active during testing. When entering test mode, there were no characterpresent, but it could be placed directly by a finger tap from the player. Thethought is that the player should not have play through the whole level everytime a part of it should be tested.

32


Figure 4.10: Springs behave like trampolines, but with a fixed amount ofbouncing force.

Figure 4.11: Start and end points.

33


Figure 4.12: Controlling the game character by tilting the iPad.

Play mode and test mode allowed the player to control a rabbit character.The available control mechanisms were:

1. unless the character was in free fall, giving the rabbit an impulse point-ing straight up, as if jumping

2. pulling the character straight left or right

There were two mutually exclusive mechanisms for pulling the character,joystick or tilting. When tilting was active, as illustrated in Figure 4.12, thevertical component of a vector pointing straight out of from the right edgeof the screen would affect the the horizontal movements of the rabbit. Tocompensate for the difference in effort required to use the two mechanisms,the tilting had a larger zone yielding 100% power than the joystick. An angleof 30◦would yield full speed when using the tilting mechanism.

There was also a joystick available when building, like the one available whentesting. There was no rabbit character to control while building, but by usingthe joystick, the player could control the panning of the camera directly.

Figure 4.13 shows the three mechanisms for moving around while editingwithout actually testing or playing.

34


Figure 4.13: Scrolling mechanisms in the game

35


Figure 4.14: The editor, when holding the iPad horizontally.

The joystick controlled the viewpoint directly by translating angle into view-point panning speed. Holding the joystick at a constant non-zero angle fromits origin would move the viewpoint at a constant speed until the limits ofthe level were reached.

The tool next to the joystick could temporarily activate a more direct viewingmechanism. While holding down one finger on the tool icon, the user coulduse finger swipes with another finger to pan the viewpoint. Translating fingerswipes into relative position changes is a common mechanism on smartphonesand post-PC tablets.

The third viewing mechanism had its own view mode. This was equivalent ofconstantly holding one finger on the swipe scrolling tool embedded in buildmode.

Figure 4.14 demonstrates that the game was agnostic about which way theiPad was held. When holding it horizontally the view rectangle wouldchange, rendering a different subset of the level. When the accelerationsensor controll mechanisms was activated this feature was deactivated forobvious reasons.

Figure 4.15 shows the play mode and the high score list mode. When playingthe levels, the GUI was similar to the one found in the editor, but withoutany editing or saving features. The character would start at the spawningpoint, which was invisible in play mode. When the character touched theflag, the elapsed time and the number of points collected would be recorded

36


Figure 4.15: Play mode and a high score list.

in the high score list of the level.

37

Chapter 5

Results and evaluation

5.1 Experience with development tools

Using the Objective-C language and the iOS, cocos2d and Box2D librariescost a lot more time than we had expected. We believe that it would be timewell spent to search for alternatives in future projects. A good game libraryshould provide all the most basic features of the UIKit, cocos2d and Box2D,eliminating the need to combine different libraries.

5.2 User tests

Procedure

The usability test was threefold; see Appendix A for the full details of thetest document.

1. Preliminary questions about the attendant’s game playing habits

2. Level building, level testing and free play as following

(a) The attendant was asked to create a simplistic level with a givenminimal amount of content. No help was given during this task.

(b) The attendant was asked to create a level containing a puzzle orother kind of challenge. No help was given during this task.

38

5.3. ANALYSIS CHAPTER 5. RESULTS AND EVALUATION

(c) The attendant was given the opportunity to do whatever he orshe wanted with the software, knowing that they were soon goingto be asked questions about it. During this part of the test theattendants were explicitly allowed to ask any kind of questions.

3. Final questions about the game and the editor - These questions in-cluded the opportunity for the attendants to give suggestions about thesoftware, written in free text.

System usability scale

The System Usability Scale can provide an assessment of the usability of asystem (Lewis and Sauro, 2009). In this project a modified version of SUSwere used. The modification was mainly to change the word ’system’ for theword ’game’. The test is shown in Appendix A and the results in Figure 5.5.

Test subjects

Test subjects were chosen from students found at the NTNU campus between9th and 11th of May. There was no explicit incentive for the students toattend the test other than the opportunity to test an iPad game for 10-15minutes and to gain a little bit of insight into the project.

42 students were asked and 33 attended the test. The majority were in their10th and final semester of a 5-year master’s programme in Computer Scienceor Communications Technology. 4 children of ages 11, 11, 14 and 15 alsoattended. The average age of the attendants was between 22 and 23 years.

5.3 Analysis

The results in Figure 5.1-5.3 are provided for reference. They reveal that thetest population consist mostly of people that play games regularly.

Figure 5.4 indicate that cellphones with keypads are unusual devices to playgames on. It also shows that post-PC tablets is far less common than cell-phones with touchscreen.

If we compare the numbers in Figure 5.4 we see post-PC tablets is the onlyplatform where more people play regularly than actually own the device.

39


20 30 40 50 60 70 80FemaleMale

% of subjects

Figure 5.1: Male and female test subjects

0 5 10 15 20 25 30 35 40

> 20h10-20h5-10h1-5h< 1h

% of subjects

Figure 5.2: Weekly time spent playing games

5 10 15 20 25 30 35

> 4h2-4h1-2h

30-60m15-30m< 15m

% of subjects

Figure 5.3: Time spent per playing session

40


Keypad cellphones have the exact opposite trend, as 5 out of 6 don’t usetheir phones for playing. Among handheld and TV console owners, one thirddon’t play.

The attendants were asked which games they had played the most the last5 years. Only 4 games were mentioned by more than 10% of the attendants,Starcraft II, World of Warcraft, Team Fortress 2 and Morrowind: Oblivion.

The results from the system usability scale gives the game a score of 74.1,this will roughly place it among the 30% most usable programs (Sauro, 2011).A score of 80.4 would place the game among the top 10% of the programs.These top score programs are more likely to be recommended by the users,which is a critical property for viral marketing.

Figure 5.6 imply that the test population feel confident about their owncreativity. It also shows, even more evidently, that the population feel com-fortable when using touch-based editors.

Figure 5.7 suggests that the test population like to be creative and sharetheir content in games, but that this is not necessarily the most importantfeature when they choose their games.

Figure 5.8 presents a population that would like to play games on the iPad,but not all agree that games with level editors are preferable.

Particularly popular and unpopular properties

The attendants were given the opportunity to give answer whether they likedor disliked anything in particular about the editor. More than 10% of theattendants gave the following, or equivalent, comments.

Good • The editor was easy to use.

Bad • I needed some more advanced objects.• I needed some explaining text in the GUI.• I (could easily have) deleted my level by accident.• I found the GUI unfamiliar.

The figures 5.9-5.12 show simplified lists of further suggestions from the testattendants. These were given as answers to the following four questions:

41


6543210

20

40

60

80

100

platform

play regularlyown device

1. Post-PC tablet

2. Cell w/ touchscreen

3. Computer

4. TV console

5. Handheld console

6. Cell w/ keypad

Figure 5.4: Test subjects who own and use devices

42


55 60 65 70 75 80 85

123456789

10

Total average

Average score

Question

Figure 5.5: System usability score (SUS)

• Were you missing anything the editor?

• Do you have any suggestions about the editor?

• Were you missing anything in the game as a whole?

• Do you have any suggestions about the game as a whole?

The results are in agreement with the first listing in Section 5.3. They suggestthree areas of improvement potential:

• There should have been more advanced game objects.

• The GUI was not intuitive enough.

• Better progress handling should have been available. This could havebeen automatic saving, undo options etc.

This suggests that the applied menu appearance strategy was a failure. Theusers were confused and did only occasionally understand the meaning ofa glyph before they had tried it out. They did, however, usually recall the

43


Strongly disagree Disagree Neutral Agree Strongly agree

0

10

20

30

40

50

I’m a creative personI feel comfortable using touch-based editors

Figure 5.6: Subjects about creativity (1/2)

44



0

10

20

30

40

I prefer games where I can make my own levelsI prefer games that have little need for creativity

I like sharing levels or other creations that I make in games

Figure 5.7: Subjects about creativity (2/2)

45



0

10

20

30

40

I like the iPad platform for playing gamesI like the iPad platform for creating game levels

Figure 5.8: Subjects about the iPad

46


1 2 3 4 5

enemiesdeath traps

scrolling toolshidden chasms

water/slowing fluidmore block types

floating pointspause/reset w/o load

undoclear all points

clear all/new levelto move objects

gravity fieldsline drawing

zoomingdangerous stuff

eventsoptions

more itemstrigger editor

iceplay mode

number of subjects

Figure 5.9: Missing tools or options in the editor

meaning of every glyph, once they had tried them at least once. Figure 5.5 didnot render the problem as critically bad, but there is room for improvement.

The figures 5.13-5.15 shows how large percentage of the population who usedsome of the different features in Platformer Playground. The attendantswere not explicitly taught how to use these features and could not ask forhelp during the first two parts of the test. The third and final part was notmandatory, but included the opportunity to ask questions and get help.

There was one attendant who did not use any of the controller-mechanismsfor the game character, but all claimed to have tested their levels and usedat least one of the scrolling mechanisms.

47


1 2 3 4 5

removable spawnpoint/goalbrush sizes

disappearing blocksmore explicit test mode

gravitation fieldice

improve delete iconimprove delete-tool icon

improve spawn/goal iconsjoystick backgroundmore detailed icons

more obstaclesok button by the text field

default goal at eastrabbit spawn icon

replace icons with textscrolling tools

zoomless powerful springs

building musicimprove view icon

quicksave/loadtutorial/info mode

number of subjects

Figure 5.10: Suggestions about the editor

48


1 2 3 4 5

enemiesmultiple saves per level

delete confirmationdiagonal elements

level restartcoop play

tilt configurationtriggers

more block typeswaterundo

death trappause button

describing text/ tutorialsharing levels

more complex levels

number of subjects

Figure 5.11: Missing options/tools in the game as a whole

49


1 2 3 4 5

automatically open new levelsguns

improve view iconinfinite chasms

more precise bounding boxesmore soundspause button

reduced spring powerremove fps counter

sharing levelsstory mode

more interaction with pointlesstilting as a global option

timer that shows while playingweaker trampoline

death trapenemies

improved iconsok button next to text field

point counterrabbit flickeringrestart button

describing text/ tutorial

number of subjects

Figure 5.12: Suggestions about the game as a whole

50 55 60 65 70 75 80

Joystick

Acceleration sensor

% of subjects

Figure 5.13: Test subjects who used controller-mechanism

50


66 68 70 72 74 76 78 80 82 84 86

Test mode

Play mode

Both modes

% of subjects

Figure 5.14: Test subjects who tested their levels

40 45 50 55 60 65 70 75 80

Dedicated swipe

Embedded swipe

Joystick scrolling

% of subjects

Figure 5.15: Test subjects who used scrolling mechanism

51



0

10

20

30

40

Improve dedicated swipe scrollingRemove dedicated swipe scrolling

Figure 5.16: Test subjects who agree to changes

When the test attendants answered whether they would suggest removing orimproving the features mentioned in the figures 5.16-5.18 the iPad and thegame was still available, and they could still ask questions.

Figure 5.16 suggests that the test attendants were happy about the dedicatedswipe scrolling mode. They do not want it improved and do not want itremoved.

Figure 5.17 is in agreement with Figure 5.15 that the swipe scrolling tool,embedded in the building mode (’/edit//build’), needed some improvement.It does, however, look like the attendants liked the tool when they finallyfound out how to use it.

It is evident from Figure 5.18 that the joystick scrolling tool was importantto the test attendants as almost 50% were in strong disagreement that thetool should be removed.

52



0

10

20

30

40

50

Improve embedded swipe scrollingRemove embedded swipe scrolling

Figure 5.17: Test subjects who agreed to changes

53


Strongly disagree Disagree Neutral Agree Strongly agree0

10

20

30

40

50

Improve joystick scrollingRemove joystick scrolling

Figure 5.18: Test subjects who agreed to changes

54



0

10

20

30

40

50

Joystick while buildingAcceleration sensor while playing already-built levels

Figure 5.19: Test subjects who prefer control mechanisms

As Figure 5.19 shows, the joystick was the preferred tool for controlling thegame character. The tilting feature appeared more like a curiosity, but waswelcomed among some users.

While the behavior of the ’point’ and ’pointless’ game objects could poten-tially increase the entropy of the game levels they were well received by thetest population. Some users mentioned that they were hard to control, sothis might mean that the entropy is not available to the users anyway. Forexample it was virtually impossible to stack more than two marbles on topof each other.

As Figure 5.20 shows, the users recognize the simplicity of the grid basedblocks better than of ’point’ and ’pointless’, but both concepts seem wellreceived.

55



0

10

20

30

40

50

Liked that ’point’ and ’pointless’ behaved physicallyLiked the simplicity of grid based blocks

Figure 5.20: Test subjects who prefer control mechanisms

56


Creative processes

We observed a lot of creativity during the tests. Appendix B shows sketchesfrom some of the game levels that the test attendants produced. The rangeof use cases included:

• simple drawings, using the grass blocks

• simple marble and spring based systems

• classic, easy platform levels

• platformer puzzles

• spring based jumping arenas

No two levels looked the same, even though they would have been easy toreproduce. All the attendants got the same tasks that they were supposedto solve, but few solved it in the same way. Most of the tests were conductedwith two attendants at the same time, and many wanted to test each othersgame levels, especially those attendants who knew each other.

Especially kids found strong motivation in the support for creative processes.A four year old girl, who did not attend the test, used the game primarilyfor drawing. She called it the "house drawing game".

57

Chapter 6

Conclusion

In this project we worked on the research questions in Section 2.3. Theserelate game level editors and the differences between game consoles, PCs andpost-PC tablets.

6.1 Post-PC tablet game level editors

General

Post-PC tablet game level editors, are much like PC game level editors andconsole game level editors. A powerful editor will create levels with muchentropy and be complex to use (El-Nasr and Smith, 2006). Users seem towant both simplicity and power, whether they are on post-PC platforms orelsewhere.

Scrolling

Scrolling is a mechanism for working with levels that are larger than a sin-gle screen. Three mechanisms were tested and they were all used by theusers. Some users only learned how to use a single mechanism, while otherenjoyed having multiple alternatives available. The most complex alterna-tive required two fingers to use and was easily misinterpreted by the usersbecause of its appearance. However, even the most complex alternative wasappreciated by the users who learned how to use it.

58

6.1. EDITORS CHAPTER 6. CONCLUSION

H1 There is no single scrolling mechanism that will fit the majority of usersin general.

This hypothesis is in agreement with the food industry’s view of consumers.There are many good recipes, and the best solution is to offer them a choicebetween the most popular ones (Gladwell, 2004).

Controlling the character in platformer games

The majority of test subjects that were given the choice between controllingthe game character with an emulated joystick or tilting the device did preferthe joystick. This result, however, is not very general, because the joystickprovided rougher precision and also much less effort than tilting the device.No experimenting with the effort and precision of the mechanisms was carriedthrough, so these differences might have had the most significant impact.

It might be best to offer the users a choice, but we would not like to makethis claim without further evidence.

Low entropy building schemes

The grid based block scheme that was used in the game was one of the mostpopular features. More than 95% of the test population reported that theywere in agreement or strong agreement that they enjoyed the simplicity ofgrid based blocks.

H2 Users generally enjoy the simplicity of low entropy building schemes.

Square buttons

(App, 2010) and (Goo, 2011) advise iOS and Android developers about thedesign of square buttons that invite users to tap them. We would furtherlike to emphasize that buttons and other widgets should be as intuitive tounderstand as possible. This advice is applicable for most platforms whereusability is a desired quality, however it requires extra attention becausethere is a convention that finger-tip-sized square buttons should be used ontouchscreen platforms.

59

6.2. SUMMARY CHAPTER 6. CONCLUSION

While some of our menu items had a satisfactory level of affordance, othersdid not. We did not investigate whether or not text should be used. Sinceour game had both buttons that were well and poorly understood, by thetest attendants, we believe that designing a comprehensive set of intuitivebuttons is possible. User tests can help assure that the designs work well forthe majority of the users.

Creativity

We believe, based on the diversity of the game levels that were producedduring our tests, that post-PC tablet game level editors can support creativeprocesses very well. We have not provided any strong evidence, but we havenot made any observations that suggest otherwise. Since creative processesare central to game level creation, this support is a prerequisite.

6.2 Summary

This project was conducted in order to answer questions about the compat-ibility between iPad, post-PC tablets, and play-create-share games.

One possible contribution towards aswering RQ1 is: approximately squareicons, that invite the user to tap them should also be self explanatory. Thisis hard to achieve and the design process should be aided with user tests andlots of continuous effort.

RQ2 has only been partly answered in this report. An emulated joystickwas the favorite choice among the test population, but the two mechanismsprovided in the game were not properly aligned for comparison. To assesswhich mechanism is the best, they should first be aligned for effort requiredand achievable precision. In the end, the users might even require bothmechanisms.

One possible answer to RQ3 is provided as H1. Editors that handle levels thatare bigger than the screen should provide a carefully selected set of scrollingmechanisms. Three examples of such mechanisms have been provided.

RQ4 has been answered only by the diversity of the games that the testattendants produced. Further investigation might reveal that better supportfor creative processes is possible. In particular, more powerful editors mighttrigger better or more complex ideas of the minds of the users.

60

6.3. FURTHER WORK CHAPTER 6. CONCLUSION

6.3 Further work

I see a bright future for play-create-share games because of all the people thatenjoyed playing the game. The implementation was small and simplistic, yetit was fun to use.

In the future I would recommend game designers to provide multiple scrollingmechanisms whenever scrolling is needed. I would also recommend puttingsome serious effort into the menu design. Buttons and similar widgets onPost-PC tablets should be more square than their PC and console counter-parts. Text litter small screens easily, so if you choose to have text, make itworthwhile.

Further research efforts could reveal better uses of the acceleration sensoror cameras. The fist model of the iPad did not have a camera, but otherpost-PC tablets and the iPad 2 do. This could open up new possibilities foreditor interfaces. For instance head tracking could be used as a mechanismfor scrolling (Bérard, 1999).

Another area that needs work is the sharing platform. I believe that nosingle, central server should be required in order to share levels with yourfriends. Central servers represent single points of failure and usually requiremaintenance (Lee, 2011). P2P technologies or local area wireless communi-cation might help remove this requirement. The XMMS or IRC protocolsor even email are communication media that could simplify the design of aserver that users would be able to set up on their own.

61

Appendix A

Test document

62

Answer the questions in the spaces provided on the question sheets. If you run out ofroom for an answer, continue on the back of the page.

Age:Sex:

About your game playing habits:

1. On an average week, how much time do you spend playing games?

A. more than 20 hours

B. between 10 and 20 hours

C. between 5 and 10 hours

D. between 1 and 5 hours

E. less than 1 hour

1.

2. On average, how long do you play one game during one game playing session?

A. more than 4 hours

B. between 2 and 4 hours

C. between 1 and 2 hours

D. between 30 and 60 minutes

E. between 15 and 30 minutes

F. less than 15 minutes

2.

Answer the following questions by putting a cross in the most appropriate box. For example, if you mean’yes’, then put a cross in the box next to ’yes’.

3. Do you play the following platform at least a few times a month?

(a) Computer

Yes No

(b) TV based console

Yes No

(c) Mobile phone with touchscreen (limited or no keypad)

Yes No

(d) Mobile phone with keypad

Yes No

(e) Post-PC tablet (iPad, Samsung Galaxy Tab, Motorola Xoom etc.)

Yes No

(f) Handheld console (PSP, DS etc.)

Yes No

4. Do have your own?

(a) Computer

Yes No

(b) TV based console

Yes No

(c) Mobile phone with touchscreen (limited or no keypad)

Yes No

(d) Mobile phone with keypad

Yes No

(e) Post-PC tablet (iPad, Samsung Galaxy Tab, Motorola Xoom etc.)

Yes No

(f) Handheld console (PSP, DS etc.)

Yes No

5. Which is your favorite platform at the moment?

5.

6. What your favourite game genre? If you have many you can list them all.

6.

7. Which platform have you played the most the last 5 years?

7.

8. Which games have you played the most the last 5 years?

Page 2

You are going to test a create-and-play gameplatform for touchscreen based handheld devices.

For us it is important to understand how well you can use the system with little or no help from us. If youhave any questions for us, you should ask them before you start testing. If the software is malfunctioninghowever, we might step in to help during the test.

You’re going to use the gameplatform to create two game levels.

• The first should be a minimalistic, but playable level containing a few grass blocks, ’pointless’ marblesand ’point’ marbles. Make sure to tell us when you think you’re done with this first level.

• The second level can be however hard you want it to be, but it should have at least one obstacle, puzzleor other kind of difficulty.

• If you complete your second level before the time limit, you may play around with the platform if youwish. Make whatever you want to make.

Page 3

Answer how much you agree to the following statements

9. I think that I would like to play this game frequently


10. I found the game unnecessarily complex


11. I thought the game was easy to use


12. I think that I would need the support of a technical person to be able to use this game


13. I found the various functions in this game were well integrated


14. I thought there was too much inconsistency in this game


15. I would imagine that most people would learn to use this game very quickly


16. I found the game very cumbersome to use


17. I felt very confident using the game


18. I needed to learn a lot of things before I could get going with this game


Page 4

19. I’m a creative person


20. I feel comfortable using touch-based editors


21. I prefer games where I can make my own levels


22. I prefer games that have little need for creativity


23. I like sharing levels or other creations that I make in games


24. I like the iPad platform for playing games


25. I like the iPad platform for creating game levels


Page 5

26. What did you like or dislike about this particular level editor?

27. There are three different scrolling tools in the level editor.

1. A dedicated view mode that use finger swipes

2. A similar finger swipe tool embeded in the building mode. To use this the player has to hold onefinger on the tool icon and make finger swipes with another.

3. A camera panning tool that uses an on-screen joystick.

(a) Did you use scrolling tool 1?

Yes No

(b) Did you use scrolling tool 2?

Yes No

(c) Did you use scrolling tool 3?

Yes No

28. Did you test your level while you were in editor mode (the mode with a checkerboard background)?

Yes No

29. Did you test your level in playmode?

Yes No

30. Did you control the character with the on-screen joystick?

Yes No

31. Did you control the character by tilting the device?

Yes No

32. Were you missing any features in the editor?

Yes No

33. If you answered yes in the previous question, what feature were you missing?

Page 6

34. Do you have other kinds of suggestions about the editor?

35. Were you missing any features in the program as a whole?

Yes No

36. If you answered yes in the previous question, what feature were you missing?

37. Do you have other kinds of suggestions about the program as a whole?

Page 7

38. I would suggest that the dedicated view mode should be improved


39. I would suggest that the dedicated view mode should be removed


40. I would suggest that the finger swipe based scrolling tool in the building mode should be improved


41. I would suggest that the finger swipe based scrolling tool in the building mode should be removed


42. I would suggest that the on-screen joystick scrolling tool in the building mode should be improved


43. I would suggest that the on-screen joystick scrolling tool in the building mode should be removed


44. I prefer controlling the character using the on-screen joystick while testing levels that I am building


45. I prefer tilting the device when playing already-built levels


46. I liked that the round game elements, called ’point ’ and ’pointless ’, behaved realistically


47. I enjoyed the simplicity of grid based grass blocks


Page 8

Appendix B

Sketches of levels created bythe test attendants

71

Appendix C

Raw answer data from thequestionnaire

75

# Age Sex 1 2 3a 3b 3c 3d 3e 3f 4a 4b 4c 4d 4e 4f1 20 m c c yes no no no no no yes no no no no no2 21 m a a yes yes yes no no no yes yes yes yes no no3 26 m e b yes no yes no no no yes no yes no no yes4 24 m c c yes yes no no no yes yes yes yes no no yes5 25 m b c yes yes no no no no yes yes no no no no6 23 m b b yes no no no no no yes no no yes no no7 24 m c b no yes yes no no no yes yes yes no no no8 23 f e d yes no yes no no no yes yes yes no no no9 24 m d c yes no yes no no no yes no yes no no no10 24 m e f no no yes no no no yes no yes no no11 23 f d e no yes yes no yes no yes yes yes no no no12 24 m b b yes yes yes no no no yes yes yes no no no13 26 m b d no no yes no no no yes yes yes no no yes14 25 m c d yes yes yes no no no yes yes yes no no no15 27 m d d yes no yes no no no yes yes yes no no yes16 25 m d f no yes yes no yes no yes yes yes no yes no17 24 m c b yes no yes no yes no yes yes no no no no18 23 m d a yes no yes no no no yes no yes no no no19 25 m c c yes yes yes no no no yes yes yes no no yes20 27 m b b yes yes yes no no no yes yes yes no no no21 23 m e e no no no no no no yes no yes no no no22 24 f d d yes yes no no no no yes yes yes no no no23 24 m c c yes no yes no no no yes no yes no no no24 23 m d b yes yes no no no no yes yes no yes no no25 24 m d b yes no yes no no no yes yes yes no no no26 23 m d c yes yes yes no no yes yes yes yes yes no yes27 19 f b b yes no yes no no yes yes no no yes no yes28 23 m c b yes yes no no no no yes yes yes no no no29 28 m e e yes no no no no no yes no no yes no no30 22 m c c yes yes yes no no no yes yes yes no no no31 28 m e b yes no no no no no yes yes yes yes no no32 23 f d b yes yes yes no no no yes yes yes no no no33 23 f b b yes yes yes yes yes yes yes yes yes yes yes yes34 11 f d d yes yes yes no yes yes no yes yes yes yes yes35 14 f d c yes yes yes no no yes yes yes yes no no yes36 11 m d d yes yes no yes no yes yes yes no yes no yes37 15 f d d no yes yes no no no yes yes yes yes yes yes

# 5 6 71 pc rpg, fps pc2 ps3 fps, action, adventure, hack 'n slash pc3 pc adventure, fps, strategy pc4 xbox360 strategy/rpg pc5 pc fps/soccer(sport) pc6 pc science-fiction pc7 xbox fps xbox8 mobile casual computer9 computer (R$s and tb-) strategy, fps, casual computer10 android Multiplayer, Rockband playstation11 iPad klikk & pek || multiplayer mariokart12 pc rollespill, strategi, casual pc13 iOS casual, rpg xbox36014 ps3 fps, strategy console15 ps3 fps, puzzle, adventure pc16 iPad Action, fps xbox17 computer puzzle, rts, rpg computer18 pc puzzles, bygge-spill, hack-n-slash RPG pc19 computer rpg computer20 computer rts, fps computer21 fps, strategy pc22 computer portal, cs ps223 pc puzzle, rpg pc24 pc strategy, mmo, fps, adventure pc25 pc rts, rpg pc26 ds adventure, rpg pc27 computer rpg, platform computer28 ps3 Platformers, Action/Adventure pc29 TV based Console sports, fighting, simple as puzzle Bubble TV console30 pc rpg, fps, brain-train, tactical pc31 pc strategy, fps, rpg, adventure pc32 mobile w/ touchscreen strategy computer33 pc adventure34 iPad platformspill, arcadespill, matspill DS, iPad, iPod, PS335 iPad Action, sport, platform xbox36 ps2 platform ps237 iPad, PS3 platform, sport, action DS

# 81 Morrowind, Oblivion2 world of warcraft, god of war franchise, red dead redemption, counter strike: source, league of legends, battlefield franchise

3 portal 1+2, HL2 + ep.1,2, CoD4, Modern Warfare 2, Monkey Island 1,2,3, Tales of Monkey Island, Psychonauts, New Super Mario Bros.(DS)

4 warcraft 3, lost odyssey5 counter strike: source, fifa 2006-011, pes 2006-2010, battlefield6 eve-online, warhammer online, total war, civilizations7 Call of Duty: black ops(multiplayer og sp), Battlefield, bad company 2 (mp og sp), tiger woods pga tour, left for dead 28 -yatzy (computer) -scrabble (mobile) -solitaire(computer)9 TF2, RockBand10 Rockband1112 oblivion13 The Elder Scrolls: Oblivion14 CoD: Modern warfare 1+2, Fifa: 09-1115 gta4, wordfeud (scrabble), assassin's creed, warlight, angry birds, team fortress16 Halo, Gears of War, WoW17 Diablo II, Starcraft (II)18 diablo 2, open transport tycoon deluxe, civilization revolution, div. Flash-spill19 World of Warcraft!, Mass effect, Saints Row 2, Final fantasy tactics A2((12)), Team Fortress 220 bad Company 2, Starcraft 2, civ. 5, Uncharted 221 starcraft, quake III22 gta23 Nurikabe, Evolution Soccer 4, mass effect, …24 WoW, SC2, CoD, FFXIII25 fallout 3, Morrowind, Oblivion26 WoW, Pokemon27 WoW, SuperMario(DS), Minecraft, Team Fortress 2, Crash Bandicoot(DS)28 Team Fortress 229 EA fifa, Angry birds, counter strike30 starcraft 1+2, minecraft, half-life 2, gta4, angry birds2, solitaire, slice it!31 Team Fortress 2, Left for dead 1+2, portal 1+2, starcraft 2, warlight, xmoto, mass effect 1+232 World of Warcraft, tower defence33 CoD, Age of Empires, SC234 Sonic Sega, allstar racing, lbp, dougnuth games35 Little Big Planet, Lego Harry Potter, Lego Star Wars3637 Little Big Planet, Sport Champion(ds move), Sega All Star racing, Lego StarWars, Lego Harry Potter

# 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 251 2 1 3 0 2 2 4 2 3 1 3 4 3 0 2 3 32 3 2 3 2 3 2 4 1 3 1 1 3 2 3 4 3 33 3 1 4 0 3 1 3 2 3 0 3 4 1 1 3 2 34 3 0 3 0 2 1 4 1 3 0 3 3 2 1 3 2 25 2 1 2 1 3 2 3 2 2 3 2 3 1 2 0 3 16 3 1 3 0 3 1 3 1 2 2 2 3 2 0 3 2 27 2 0 3 0 4 0 3 0 3 0 3 4 2 2 3 3 38 3 3 2 1 3 1 3 2 2 1 2 4 3 1 3 4 49 1 0 3 0 3 1 3 1 1 0 2 3 1 2 1 2 210 0 1 3 0 4 1 4 0 3 0 2 3 2 1 3 4 411 2 1 3 1 3 0 4 0 3 0 1 4 1 3 3 4 412 3 0 4 1 3 0 4 0 4 0 3 3 3 0 2 3 313 3 1 3 0 4 1 4 0 3 0 3 4 3 0 3 4 314 0 1 2 0 3 2 3 1 2 0 3 3 1 2 1 3 215 1 1 2 1 3 1 2 3 2 1 2 4 2 1 2 3 316 2 1 2 1 2 1 3 1 2 1 3 3 1 1 1 3 217 2 1 3 1 2 1 3 2 2 1 3 4 3 1 4 4 318 4 0 4 0 4 0 4 1 4 0 4 4 2 2 1 4 319 2 1 3 0 3 1 4 0 1 1 2 3 1 2 1 3 320 2 1 3 0 3 0 3 1 4 0 2 2 4 1 2 3 321 2 0 3 1 4 1 1 0 3 1 4 4 3 0 3 2 322 3 1 1 1 1 2 3 3 1 1 3 1 3 1 1 3 123 1 1 3 1 3 2 3 1 3 1 2 3 1 3 1 3 224 2 1 3 0 2 1 3 2 3 1 1 2 2 2 2 2 225 3 1 3 1 0 0 3 1 3 0 3 3 2 0 3 2 226 2 0 3 1 3 1 4 1 2 1 3 4 1 2 1 3 227 3 1 3 0 4 0 3 3 0 2 4 2 1 3 2 228 2 0 4 0 3 0 4 0 4 0 2 3 2 2 3 2 229 2 0 3 1 3 0 2 1 3 1 3 3 2 3 2 3 330 3 1 3 1 2 2 4 0 2 0 2 4 4 0 4 4 431 0 0 3 0 3 1 4 0 1 0 4 4 2 0 2 2 332 3 1 3 1 3 1 4 0 2 1 2 3 3 1 4 4 433 3 3 1 2 1 2 1 3 1 2 3 4 2 4 3 3 334 2 0 3 1 2 2 4 0 4 0 4 3 4 1 3 4 435 2 1 3 0 2 2 2 0 4 0 3 4 2 2 2 4 436 4 0 4 0 4 0 4 0 4 0 3 4 4 4 4 4 437 2 1 3 1 3 1 2 1 2 1 3 4 3 1 3 3 3

# 26

1 Ikke helt intuitivt hvordan man lagrer og spiller brettet med en gang, kunne vært muligheter for mer avanserte leveler. Men gikk fort å finne ut hvordan alt fungerer.

2 no "move object function". Easy to select objects. Good overview. No zoom buttons3 Cumbersome: -placing balls/points -deleting Missed textual icons4 like: easy to use5 Some levels where not the same when editing as in gameplay. Lower grass was not visible f.eg.

6like: simplicity, but option to make complex stuff with it; dislike: more fancy features

7+likte gui-et veldig godt +morsomt konsept -det jeg "gjorde" (f.eks plukka poeng) i testmodusen til editoren ble værende. -

savnet pinch-to-zoom når jeg så meg rundt -hadde likt uendelig eller betydelig mer plass til å bygge på. -flere "hindere" -skulle gjerne valgt om pointsa skulle stå stille eller bevege seg i edit mode

8

touch-screen gjorde det enkelt å lage levels; enkel funksjonalitet

9 +easy to use +clean looking +many possibilities -could have been more explanations available. E.g. Difference between point marbles and pointless marbles

10 var noen veldig flotte gni-funksjoner som var behagelig å bruke; var lystbetont spill

11 tok litt tid å finne lagre-funksjonen

12 Veldig lett å teste brett man utvikler, raskt og effektivt.; hadde litt problemer med menyen intill jeg skjønte fargemarkeringen

13 kunne brukt tekst i tilleg til ikoner for å vise hva elementene gjør

14 -pointless is very pointless -not a lot of options +easy to start and test

15 *To easy to mess up (delete/destroy contraption) *Not always intuitive

16 Lite hjelp: informasjon

# 2617 styre rundt om man skulle utafor en skjerms størrelse18 Mystery Meat Navigation; "ok"-tast i menyene i stedet for enter; restart-button trengs; musikk i editor trengs19 Litt upraktisk skjermscrolling

20 Likte: God oversikt over brettet. De fleste ikoner lett å skjønne. Mislikte: Endte opp med å slette et brett da jeg trodde deleteknappen var cancel og gå tilbake.

21 -scrolling med to fingre -blinking av kaninen22 lite intuitiv, men lett når man skjønte hvordan23 Does not seem like it can offer enough variety for now24 Like: easy to use, short amount of time needed to play; dislike: few options, I have to be very creative to make a cool game25 kanskje litt begrensede artifakter26 I liked the simplicity. Took some time to understand and recognize some of the tool glyphs. Struggled with delete.27 I liked the simplicity, …28 tooltip for the icons29 the icons could be confusing30 dislike: preview, a little bit few things31 flytte tiles

32 Enkel å sette seg inn i. Ikonene ga mening. Litt dumt at man ikke visste f.eks hvor høyt kaninen hoppet og hvordan gjenstandene virket. Men dette skjønner man fort når man spiller spillet.

33

34

35

36

37

# 27a 27b 27c 28 29 30 31 321 yes yes no yes yes no yes no2 no no yes no yes yes yes yes3 yes no yes yes yes yes no yes4 yes no yes yes yes yes no yes5 yes no no yes no yes no no6 yes no yes yes yes yes no yes7 yes yes no yes no no yes no8 yes no yes no yes yes no no9 no no yes yes yes yes no no10 yes yes no no yes yes yes no11 no no yes yes yes yes no yes12 yes yes yes yes yes no yes yes13 yes no yes yes yes yes no no14 no no yes yes yes yes no yes15 yes yes yes yes yes yes yes yes16 no no yes yes yes yes no no17 no no yes yes yes yes no no18 yes yes yes yes yes yes yes no19 yes no no yes yes no yes yes20 yes no yes yes yes yes no no21 yes no yes yes yes yes yes yes22 yes no no no yes no no no23 yes no no yes no yes no no24 yes no yes yes yes yes yes yes25 yes yes yes yes yes yes yes yes26 no no yes no yes yes yes no27 yes yes no yes no no yes no28 yes no no yes yes yes no no29 yes yes no yes yes yes yes no30 yes yes yes yes yes yes yes yes31 yes no yes yes no yes yes no32 yes yes no no yes no yes no33 yes yes yes yes yes yes no yes34 yes yes yes no yes yes no no35 yes yes yes yes yes yes yes yes36 no no yes yes yes yes no no37 yes yes yes yes yes yes yes yes

# 3312 scrolling tools3 clear all poits, clear all/new level4 to be able to move obstacles that are already placed56 gravity fields/ hover fields, death traps, chasms(hidden), water, etc.7891011 *flere typer blokker *kunne dra en hel linje (ble litt hullete)12 fiender1314 enemies or floating points15 Pause/reset (w/o explicit load)16171819 Litt flere platformelementer. Fiender, fallgruver, etc.; Undo-funksjonalitet; Multitouch scrolling/zooming2021 poeng"kuler" som ikke beveger seg; farlige ting222324 more items, events, options25 trigger editor2627282930 trap, glas/ice, slowing fluid, enemies313233 play mode3435 Jeg savnet "monster" som du mister liv av.3637 undo/redo button, dødelige hinder, av/på-knapper

# 3412 zoom and move mode/button as aforementioned3 some nice building music maybe4 different "brush"sizes. To enable placing several tiles at the same time56 quicksave/quickload7 flere hindere. For eksempel dings som flipper om gravitasjonen osv.8 nederste ikon var vanskelig å forstå9 Maybe a "intro" feature for explaining things. Perhaps a "?" mark "mode".101112 is for å få kaninen til å skli, bakker som forsvinner over tid13 konsistens ved sletting av start/slutt i likhet med andre elementer me slett-knappen1415 *paint bak joystick? *rar startpos for start/mål *confirm delete (yes/no) *kryss-ikon vanskelig å forstå16 Noen av ikonene passet ikke den funksjonen de hadde. Litt vel simplistiske ikoner1718 sp.m. 261920 Under verktøy-fanen, bruk navn i stedet for ikoner. (save, load, delete, osv.)21 plasserig av flagget helt til høyre2223242526 some of the glyphs could be a bit more clear. Perhaps a tutorial mode?27282930 preview needs to be reset, points are dissappearing from the editor31 pinch to zoom3233 *en pil eller henvisning til at det er en kanin *flagg=mål *mann=start *mer beskrivende at man er I buildmode og ikke I playmode?3435 Noe annet enn et øye "to scroll"3637 kanskje en hånd I stedet for øye "to scroll" og trampolinen hadde kanskje litt for stor sprett

# 35 361 yes mulighet for mer komplekse leveler2 yes text3 no4 no5 yes to add enemies / be able to die; enemies would be both static and moving6 yes savegame management, option to save/load to/from a list of savegames for each level.7 yes -en "are you sure you want to delete" -mulighet til å dele brettet med venner/omverdenen8 yes mulighet for å fjerne elementer ((nederste ikon))9 yes An easy way of sharing levels with friends. It's more fun when you can share levels.10 yes mulighet for skråplan i gresset11 yes mini-tutorial / eller små piler som sier -> lagring her el.12 no13 no14 no15 yes Level restart16 no17 no18 yes coop play, knapper for å åpne dører19 yes Konfigurerbar respons på helling. Default som krever ganske stor vinkel for å få kaninen opp i god fart20 no21 no22 no23 no24 no25 yes triggers26 no27 no28 no29 yes more building blocks to make more complex levels30 yes more things to do31 no32 no33 no34 yes flere bakketyper, vann?35 yes Monster, forskjellige "bakker" (ikke bare gress).36 no37 yes undo knapp, pigger/dødelige hinder, evt. Av/på-knapp

# 37

1

2 more text explaining the different buttons and better score system3 it's a bit simple more features on a story/fixed levels would be nice probably4 the rabbit's movement was a bit unstable when jumping alongside a wall.5 add a help menu, f.eg. To explain the view modes in editor

6

7

8

*litt vanskelig å forsta forskjell på ikon som delete/load/save/exit. Burde enten hatt tydeligere ikoner eller hatt tekst ved siden av. (spelielt load/save) *Synes ikke øyet var et veldig tydelig symbol, men noe man lett forstår etter å ha testet

funksjonaliteten en gang * samle ikonene i toppen av skjermbildet -her er det mye ledig plass. kom ved flere tilfeller borti ikonene når jeg skjulle bygge * help kunne vært fint å ha med i hovedmenyen. Evt. Instruction *Hadde litt problemer med å

lage ny playground. Ikke intuitivt å måtte trykke på return

9

10 tidtaker som vises mens man spiller

11

12

13

14 overpowered trampolin; its possible to get stuck, thats no good; should be able to restart

15

16

# 371718 -Fikse flickering -bounding boxes er litt for store i forhold til sprites, så man kan stå utenfor kantene19

20power-ups; dører+nøkler

21 Huske at man har trykket på ((tilte-symbol))22 kunne vært enkle instrukser om bruk. Burde også vært lettere å starte spillet.2324 After writing the name of the new level, one should auto step into the game2526 Ability to download maps from other players.27282930 easier buttons; hard to understand view and preview; remove fps counter :-p31 aliens og skytere

32

33*mer beskrivende ikoner? *forskjellig lyd til byggetingene *flagg="mål"-lyd *start="start"-lyd ol.

34 du får vite hvor mange poeng du har fått

35 Trampolina skulle kanskje ikke hoppe så høyt. (Ha flere muligheter med høyde) Miste live hvis man detter ned I hull eller blir tatt av monster. Poengsum bør vises. "Start over"-knapp.

36 At vi kunne spille online og multiplayer med andre

37noe mer å bruke pointless-greiene til, evt. Til å skru av/på en mekanisme som f.eks flyttbare platformer. En "start over"-

knapp I spillemodus I tilfelle man blir sittende fast. Litt dramatisk musikk til å være stort sett uten farer? Miste liv/"dø" hvis man detter ned der det ikke er gress. Siden man kan få tak i poeng, bør det vises hvor mye poeng(evt. av x mulige) når

banen er fullført.

# 38 39 40 41 42 43 44 45 46 471 2 0 2 0 2 0 1 1 4 42 2 1 3 2 1 1 3 2 3 43 1 1 2 2 1 0 3 2 3 44 1 2 2 2 3 1 4 2 4 45 3 1 2 0 1 2 3 1 3 46 1 0 2 2 2 2 4 1 3 37 3 0 3 0 2 3 1 3 4 48 3 1 1 0 3 2 2 4 4 49 0 1 1 1 1 1 3 1 3 310 1 0 2 0 2 0 1 1 3 411 2 4 2 1 3 0 4 1 1 412 0 2 2 3 0 0 2 4 2 413 0 0 3 1 0 0 4 2 3 414 1 3 2 1 3 1 3 1 2 315 0 1 2 0 3 0 3 0 3 316 2 2 2 2 1 1 3 1 3 317 0 3 4 0 0 4 4 0 4 418 2 4 3 0 4 0 4 0 3 419 3 0 4 0 1 1 2 3 3 320 0 1 2 0 0 0 2 1 3 321 2 2 4 2 1 2 2 3 4 322 3 2 2 2 2 2 2 3 3 323 2 2 2 2 2 2 3 2 2 324 1 0 1 0 1 0 2 2 3 325 2 1 2 1 3 0 3 1 3 326 2 2 2 1 1 0 3 2 4 427 1 1 1 0 3 1 1 3 3 428 1 2 1 0 1 0 3 1 3 329 2 2 3 2 1 0 3 1 3 330 0 1 4 2 0 0 4 0 4 431 0 3 2 1 2 0 3 0 4 232 1 1 3 0 2 2 2 3 3 433 1 1 2 1 2 1 3 3 1 334 2 2 2 2 3 2 3 0 2 335 2 1 3 1 2 1 3 2 4 336 0 1 0 0 4 0 4 2 4 437 1 0 1 0 1 0 3 3 3 4

References

Answerbag. Definition of Platform game. http://www.answerbag.com/t_view/269, 2003. URL http://web.archive.org/web/20070211211612/http://www.answerbag.com/t_view/269. [Online; accessed 11-Feb-2007].

iPad - Technical Specifications. Apple, 2009. URL http://liveweb.archive.org/http://support.apple.com/kb/SP580.

Custom Icon and Image Creation Guidelines. Apple, 2010. URLhttp://web.archive.org/web/20101203210326/http://developer.apple.com/library/ios/#documentation/userexperience/conceptual/mobilehig/IconsImages/IconsImages.html.

Victor Basili. The experimental paradigm in software engineering. InH. Rombach, Victor Basili, and Richard Selby, editors, Experimental Soft-ware Engineering Issues: Critical Assessment and Future Directions, vol-ume 706 of Lecture Notes in Computer Science, pages 1–12. Springer Berlin/ Heidelberg, 1993. URL http://dx.doi.org/10.1007/3-540-57092-6_91.

Jonathan Blow. Game development: Harder than you think. Queue, 1:28–37,February 2004. ISSN 1542-7730. doi: http://doi.acm.org/10.1145/971564.971590. URL http://doi.acm.org/10.1145/971564.971590.

François Bérard. The perceptual window: Head motion as a new inputstream. In Proc. Interact99, Edinburgh, A. Sasse &; C. Johnson Eds,IFIP IOS Press Publ, 1999. 238 pages.

David Buckingham and Andrew Burn. Game literacy in theory and practice,2007.

Kate Compton and Michael Mateas. Procedural level design for platformgames. In Proceedings of the 2nd Artificial Intelligence and Interactive

89

http://www.answerbag.com/t_view/269

http://www.answerbag.com/t_view/269

http://web.archive.org/web/20070211211612/http://www.answerbag.com/t_view/269

http://web.archive.org/web/20070211211612/http://www.answerbag.com/t_view/269

http://liveweb.archive.org/http://support.apple.com/kb/SP580

http://liveweb.archive.org/http://support.apple.com/kb/SP580

http://web.archive.org/web/20101203210326/http://developer.apple.com/library/ios/#documentation/userexperience/conceptual/mobilehig/IconsImages/IconsImages.html



http://dx.doi.org/10.1007/3-540-57092-6_91

http://dx.doi.org/10.1007/3-540-57092-6_91

http://doi.acm.org/10.1145/971564.971590

REFERENCES REFERENCES

Digital Entertainment Conference (AIIDE), pages 109–111, Marina delRey, California, June 2006.

Magy Seif El-Nasr and Brian K. Smith. Learning through game modding.Comput. Entertain., 4, January 2006. ISSN 1544-3574. doi: http://doi.acm.org/10.1145/1111293.1111301. URL http://doi.acm.org/10.1145/1111293.1111301.

Box2D Physics Engine. Erin Catto and the Box2D community, 2011. URLhttp://web.archive.org/web/20090630020943/http://box2d.org/.

Joachim Froholt. Intervju: Minecraft og fremtiden. http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/, 2010.URL http://liveweb.archive.org/http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/. [Online; accessed 9-June-2011].

Malcom Gladwell. Malcolm Gladwell on spaghettisauce. http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software/, 2004. [Online; accessed 6-June-2011].

Icon Design Guidelines, Android 2.0. Google, 2011. URLhttp://web.archive.org/web/20101203061340/http://developer.android.com/guide/practices/ui_guidelines/icon_design.html.

Sal Humphreys, Brian Fitzgerald, John Banks, and Nic Suzor. Fan-based pro-duction for computer games: user-led innovation, the “drift of value” andintellectual property rights. Media international Australia incorporatingculture policy, 114:16–29, 2005. URL http://search2.scholarsportal.info.myaccess.library.utoronto.ca/ids70/view_record.php?id=1&recnum=22&SID=760717fafbf1a1c5a4539da053640d93.

Garnett Lee. Sony confirms PlayStation Networkhacked. http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked/, 2011. URLhttp://liveweb.archive.org/http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked. [Online; ac-cessed 9-June-2011].

James Lewis and Jeff Sauro. The factor structure of the system usabilityscale. In Masaaki Kurosu, editor, Human Centered Design, volume 5619

90

http://doi.acm.org/10.1145/1111293.1111301

http://doi.acm.org/10.1145/1111293.1111301

http://web.archive.org/web/20090630020943/http://box2d.org/

http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/

http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/

http://liveweb.archive.org/http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/

http://liveweb.archive.org/http://www.gamer.no/artikler/73142/intervju-minecraft-og-fremtiden/

http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software/

http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software/

http://web.archive.org/web/20101203061340/http://developer.android.com/guide/practices/ui_guidelines/icon_design.html

http://web.archive.org/web/20101203061340/http://developer.android.com/guide/practices/ui_guidelines/icon_design.html

http://search2.scholarsportal.info.myaccess.library.utoronto.ca/ids70/view_record.php?id=1&recnum=22&SID=760717fafbf1a1c5a4539da053640d93



http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked/

http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked/

http://liveweb.archive.org/http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked

http://liveweb.archive.org/http://www.shacknews.com/article/68215/sony-confirms-playstation-network-hacked

of Lecture Notes in Computer Science, pages 94–103. Springer Berlin / Hei-delberg, 2009. URL http://dx.doi.org/10.1007/978-3-642-02806-9_12.

Andrzej Lingas. The power of non-rectilinear holes. In Proceedings of the 9thColloquium on Automata, Languages and Programming, pages 369–383,London, UK, 1982. Springer-Verlag. ISBN 3-540-11576-5. URL http://portal.acm.org/citation.cfm?id=646236.682869.

Simon Egenfeldt Nielsen, Jonas Heide Smith, and Susana Pajares Tosca.Understanding Video Games: The Essential Introduction. Routledge, newedition edition, February 2008. ISBN 0415977215. URL http://www.worldcat.org/isbn/0415977215.

St.meld. nr. 14 (2007-2008), Dataspill, 2007-2008. The norwegian ministryof culture.

Kumar Sanjeev, Chhugani Jatin, Changkyu Kim, Daehyun Kim, NguyenAnthony, Dubey Pradeep, Bienia Christian, and Youngmin Kim. Secondlife and the new generation of virtual worlds. Computer, 41(9):46 –53,sept. 2008. ISSN 0018-9162. doi: 10.1109/MC.2008.398.

Jeff Sauro. Measuring Usability - Quantitative Usability, Statistics &Six Sigma. 2011. URL http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software. [Online; accessed 6-June-2011].

Noor Shaker, Georgios Yannakakis, and Julian Togelius. Towards automaticpersonalized content generation for platform games.

Penelope Sweetser and Peta Wyeth. GameFlow: A Model for EvaluatingPlayer Enjoyment in Games. The University of Queensland, St Lucia, Aus-tralia, 2005. URL http://portal.acm.org/citation.cfm?id=1077253.

Gregory Taefay. Casual Game Design: Designing Play for the Gamer in Allof Us, 2010.

cocos2d for iPhone. Zynga and the cocos2d community, 2011.URL http://web.archive.org/web/20100621051008/http://www.cocos2d-iphone.org/.

http://dx.doi.org/10.1007/978-3-642-02806-9_12

http://dx.doi.org/10.1007/978-3-642-02806-9_12

http://portal.acm.org/citation.cfm?id=646236.682869

http://portal.acm.org/citation.cfm?id=646236.682869

http://www.worldcat.org/isbn/0415977215

http://www.worldcat.org/isbn/0415977215

http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software

http://aiweb.techfak.uni-bielefeld.de/content/bworld-robot-control-software

http://portal.acm.org/citation.cfm?id=1077253

http://web.archive.org/web/20100621051008/http://www.cocos2d-iphone.org/

http://web.archive.org/web/20100621051008/http://www.cocos2d-iphone.org/

play-create-share gameplatform for the ipadthe ipad’s touchscreen and acceleration sensor. we...

Documents